Development and Early Adoption of Electric Vehicles: Understanding the tempest

(1)

William S. Sierzchula Delft University of Technology

(2)

(3)

Proefschrift

ter verkrijging van de graad van doctor aan de Technische Universiteit Delft,

op gezag van de Rector Magnificus prof. ir. K.C.A.M. Luyben, voorzitten van het College voor Promoties

in het openbaar te verdedigen op maandag, 19 januari 2015 om 12:30 uur

door

William Stanley SIERZCHULA

Master of Public Affiars and Urban & Regional Planning geboren te Fayetteville, Arkansas, Verenigde Staten

(4)

(5)

Dit proefschrift is goedgekeurd door de promotor: Prof. dr. G.P. van Wee

Copromotor: dr. K. Maat

Samenstelling promotiecommissie:

Rector Magnificus Voorzitter

Prof. dr. G.P. van Wee Technische Universiteit Delft, Promotor dr. K. Maat Technische Universiteit Delft, Copromotor Prof. dr. ir. Bart van Arem Technische Universiteit Delft

Prof. dr. Ans Kolk Universiteit van Amsterdam Prof. dr. Cees van Beers Technische Universiteit Delft

dr. ir. R. Smokers TNO

Prof. dr. ir. Paulien Herder Technische Universiteit Delft

TRAIL Thesis Series T2015/2, the Netherlands TRAIL Research School TRAIL Research School

PO Box 5017 2600 GA Delft The Netherlands T: +31 (0) 15 278 6046 F: +31 (0) 15 278 4333 ISBN: 978-90-5584-177-6

Keywords: electric vehicle, eco-innovation, policy, technological change Cover illustration: The teaser image of the Tesla Model X

All rights reserved. No part of the material protected by this copyright notice may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying, recording or by any information storage retrieval system, without written permission from the author

(6)

(7)

i

Preface

I am pleased to present this thesis as a result of my PhD research at the Delft University of Technology.

A quick reference to a no-less-venerable source than Wikipedia tells me that a preface “generally covers the story of how the book came into being . . . often followed by thanks and acknowledgments to people who were helpful to the author during the time of writing.” Well, here goes.

After following the lovely and talented Brynne DeNeen to the Netherlands in the winter of 2008, I worked for the better part of a year as a junior policy advisor at the Dutch social housing corporation WonenBreburg. Thereafter, I found myself in the seemingly enviable position of being a gentleman of leisure. However such a life did not suit me, and I quickly became restless, leading to a lengthy and remarkably unsuccessful job search. One kind gentleman at the Delft University of Technology offered to meet and talk about academic employment. Bert van Wee humored me by having the discussion in Dutch, and suggested I apply for an upcoming PhD opening looking at electric vehicles, specifically “The environment of early adopters from an innovations perspective”. To my (and many others) great relief, I was awarded the appointment, and thus began four years of academic wandering, dealing with the inevitable vagaries that accompany research in a new discipline. The task of writing a PhD thesis entails mastering a substantial amount of material. I needed to obtain a firm grasp of the relevant theory, empirical literature, and research methods in an area in which I was decidedly unfamiliar – the development and early adoption of electric vehicles. Thankfully, for this task I had the help of my three intrepid advisors.

To Bert van Wee, Sjoerd Bakker, and Kees Maat, this thesis would not have been possible without your steadfast support and guidance. Bert, your thoughtful advice along with remarkably fast responses to e-mails has always kept me (more-or-less) properly focused during my study. Sjoerd, thank you for infusing this thesis with more than a touch of innovation studies while also greatly influencing my personal and professional interest in technological change. And to Kees, your focus on scientific rigor has given me an appreciation for how research should be properly done. In addition, your push for me to engage the literature, identify a worthwhile research question, and present my analysis in

(8)

clear manner has had a tremendous and positive impact on each of the studies in this volume. It has also led to my greater appreciation for science-based evidence and made me increasingly skeptical of arguments made in the popular media, a quality I hope many others share.

And finally, to my family and friends, it is impossible for me to express how thankful I am to have such wonderful people in my life. I love you all very much.

Will Sierzchula

(9)

iii

L

ist

of Figures

Figure 1-1: Powertrain innovations relative to the ICE and fueling infrastructure ... 4

Figure 1-2: Thesis chapter positioning within the broader EV industry ... 12

Figure 2-1: Sample of alliances for BMW ... 29

Figure 2-2: Distribution of explorative and exploitative alliances by year ... 32

Figure 2-3: Explorative and exploitative alliances in key knowledge areas ... 34

Figure 2-4: Explorative and exploitative alliances in key knowledge areas by firm ... 35

Figure 3-1: Powertrain innovations relative to the powertrain and fueling infrastructure ... 48

Figure 3-2: Number of AFV models introduced ... 56

Figure 3-3: Average number of different AFV technologies presented by manufacturers ... 56

Figure 3-4: Number of firms that introduced an AFV model ... 56

Figure 3-5: Moving 3 year average of AFV production models from 1991 to 2011 ... 57

Figure 3-6: Moving 3 year average of AFV prototype models from 1991 to 2011 ... 57

Figure 3-7: Electric vehicle models ... 59

Figure 3-8: Hybrid-electric vehicle models ... 59

Figure 3-9: Hydrogen vehicle models... 60

Figure 3-10: CNG vehicle models ... 60

Figure 3-11: LPG vehicle models Figure 3-12: Flex-fuel vehicle model ... 61

Figure 3-13: Number and type of AFV models ... 62

Figure 4-1: Research framework for eco-innovations during an era of ferment ... 76

Figure 4-2: Companies producing electric vehicles ... 79

Figure 4-3: Unique battery chemistries in electric vehicle models ... 81

Figure 4-4: Electric vehicle classification by manufacturer type ... 82

Figure 4-5: 2008 German and United Kingdom new car registrations ... 83

Figure 4-6: Electric vehicles by top speed and manufacturer type ... 84

Figure 4-7: Conclusions represented in the analytical framework ... 85

Figure 5-1: Financial incentives by country and corresponding EV market share ... 102

Figure 5-2: Breakdown of financial subsidies types offered by countries ... 103

Figure 5-3: National charging infrastructure and EV market share by country ... 104

Figure 5-4: Number of EV models available for purchase, production facilities, and national market shares ... 105

(14)

(15)

ix

List of Tables

Table 1-1: Characteristics of midsize alternative fuel vehicles ... 4

Table 1-2: Role of actors relative to barriers to the development and adoption of EVs... 8

Table 2-1: EV inter-firm network by number of firms and alliances ... 31

Table 2-2: P-value from paired-sample t-tests in key knowledge areas ... 33

Table 2-3: P-value from binomial analysis of variance in key knowledge areas ... 33

Table 2-4: Average number of explorative and exploitative alliances per firm type by year . 36 Table 3-1: Fuels, barriers, and advantages of alternative fuel powertrains ... 49

Table 3-2: Transportation policies ... 50

Table 3-3: Vehicles sold, leased, or converted in the US from 2000-2009 ... 52

Table 3-4: Vehicles produced in Japan from 2000-2009 by powertrain type ... 52

Table 3-5: AFVs prototype or production status ... 56

Table 4-1: Electric vehicles that were sold, leased or converted ... 77

Table 4-2: Vehicle classification scheme ... 78

Table 4-3: Vehicle classification scheme ... 78

Table 4-4: Lithium-ion battery chemistries ... 81

Table 5-1: Description of variables and sources ... 99

Table 5-2: ICE vehicle and electric vehicle used for policy valuation ... 100

Table 5-3: Regression results for 2012 electric vehicle adoption ... 106

Table 5-4: Model sensitivity analyses 1 and 2 ... 108

Table 5-5: Model sensitivity analyses 3–5 ... 109

Table 6-1: Overview of study sample ... 122

Table 6-2: Factors that influenced fleet managers’ initial adoption of electric vehicles ... 123

Table 6-3: Exemplary quotes from textual categories ... 124

Table 6-4: Influential factors for organizations that expanded their EV fleets ... 125

(16)

(17)

1

1. Introduction

The storm is up, and all is on the hazard. William Shakespeare

1.1. Uncertainty in the automotive sector

Due to factors such as climate change, dependence on unpredictable autocratic regimes for fuel, and depletion of finite oil resources, a vast transitional period in automobility could be underway. The drive to reduce the amount of oil that automobiles use and the greenhouse gases (GHG) that they give off has resulted in the creation of new technologies and implementation of stringent emissions regulation. Manufacturers have responded in a variety of ways including improving the efficiency of internal combustion engines and developing vehicles that use alternative fuels e.g., hydrogen and compressed natural gas (Yeh et al., 2007; Oltra and St. Jean, 2009; Yu et al., 2010; Bakker et al., 2012). In particular, electric vehicles (EVs) are seen as being one of the most promising innovations to reduce oil usage and GHG emissions from the transportation sector because they do not require gasoline/diesel for operation, there is a broad existing electricity infrastructure, and many firms have already commercialized production models (IEA, 2013). However, these activities have led to an increase in uncertainty regarding how the automobile industry will address oil and climate change issues, and more specifically the role of EVs therein. In order to provide insight into the situation, this thesis seeks to understand the dynamics which underpin EV development and market introduction.

Before discussing the specific research aim and question of this thesis, an overview is provided of topics that are relevant for the success or failure of EVs. This will help to understand gaps in the literature and consequently the PhD thesis’s research aim and contribution. The following Section (1.2) describes the basic technical characteristics of EVs and compares them to other alternative fuel vehicles (AFVs) and an internal combustion vehicle (ICEV). Section 1.3 introduces the theoretical barriers that limit development and adoption of EVs. Section 1.4 identifies actors which play important roles during the emergence of EVs. Section 1.5 combines the technical characteristics, theoretical barriers, and important actors from the previous three sections in a brief literature review of the history of EVs from 1990-2010. Thereafter, Section 1.6 identifies the literature gap along with the

(18)

analyses that will address the associated research question. Finally, Section 1.7 addresses research scope, data, and methods.

1.2. Technical overview of electric vehicles

The technical characteristics of EVs provide a basis for understanding potential environmental gains and also the barriers that limit their development and wide-spread adoption. The following section provides an overview of additional important technological aspects including the battery, driving range, and refueling infrastructure. In addition, Table 1-1 also identifies how EVs relate to other AFVs by comparing the automobiles’ basic features. 1.2.1. Vehicle emissions

While in many cases EVs produce fewer pollutants that ICEVs, this is dependent on several factors, primarily the source fuel of their electricity. When using an environmental life cycle assessment, EVs provide 10%-24% lower levels of GHG emissions (based on the present European electricity mix) than a comparable ICEV, although the precise ratio is dependent on the power grid mix, speed and load conditions, and vehicle lifetime in kilometers (Hawkins et al., 2012; Ma et al., 2012). In areas where electricity is primarily produced by coal plants (such as China), EVs emit on average 3.6 times as much hazardous particulate matter1 than gas-powered ICEVs2 (Shuguang et al., 2012). But since EVs do not necessarily use a carbon-based fuel, theoretically their pollution emissions would be extremely low if electricity comes from a clean source such as solar or wind. Operationally, this number can be zero, but some carbon would still be needed for production and disassembly. In addition, by not having tailpipe emissions, EVs provide localized environmental benefits through lower particulate matter levels, NOx, and noise pollution (Shuguang et al., 2012). And as a country’s energy production shifts from coal to nuclear, gas, and renewables, these environmental benefits become more pronounced.

1.2.2. Batteries, price, and range

Historically, EV powertrains have used a variety of different battery chemistries including Nickel Metal Hydride, Lithium-ion (Li-ion), Lead-Acid, and Sodium-Nickel-Chloride. Low-speed EVs3 generally use lead-acid batteries while EVs that are similar in size/speed to conventional automobiles use Li-ion batteries (ITAQ, 2008; Lowe et al., 2010). Due to their high cost per kilowatt hour (kWh), Li-ion batteries greatly influence both the purchase price and driving range of EVs. Most ‘high-speed’ EVs (which will be the focus of the majority of this thesis) cost between $30,000 and $40,0004, and have a 75-100 mile range e.g., Nissan Leaf, Ford Focus EV, and Honda Fit EV (Autotrader, 2013), although the $70,000 Tesla Model S that goes 200 miles on a single charge5 (Tesla, 2013) does show how additional kWh’s improve performance. As of the writing of this thesis (early 2014), no company had produced a mass market EV with a driving range equivalent to a comparable ICEV. This is likely because the vehicle cost would be so high that it would only be appealing to a niche market e.g. the Tesla Model S. Because battery costs display such a powerful influence in increasing vehicle prices, they are considered to be the most important factor limiting EV adoption (IEA, 2011; Wells and Nieuwenhuis, 2012). Improvements in battery prices have

1

PM2.5

2_{This is 2.5 times higher than diesel powered ICEVs.}

3_{These are small EVs with top speeds below 25 miles per hour.} 4

In 2014 and not including federal/state rebates.

5_{There are three battery options for the Tesla model S, 60kWh, 85 kWh, and 85 kWh performance. The}

(19)

progressed slowly even though auto manufacturers and power storage firms have been spending billions of dollars developing new technologies over the past several decades (Dijk et al., 2013). While some research expects that EV battery costs will dramatically decrease in the future, it is worth noting that such price reduction expectations have not been met in the past (Bedsworth and Taylor, 2007).

1.2.3. Charging and infrastructure

EVs require up to several (>10) hours from a 110 or 220 volt outlet or approximately 30 minutes using a fast charging station, dependent on battery size (Saxton, 2013). This represents a significantly longer period than the standard four minutes necessary to fuel an ICEV, contributing to a negative association of EVs by consumers. Furthermore, with fast charging there is the possibility that it might have a detrimental effect on a battery’s energy density after repeated use (Boulanger et al., 2011).

Regarding the power grid, EVs represent both an opportunity to improve load balancing, but also the potential to intensify existing uneven energy demand cycles. The daily energy system load sees electricity demand ramp up between the hours of 5:00am and 8:00am, remain roughly level throughout the workday, peak between 4:00pm and 7:00pm, and then trail off. Such variety in usage entails a high capacity level, in accordance with peak demand, that is not utilized throughout much of the day (Lemoine et al., 2008). This is an inefficient setup which requires some power plants to rapidly increase their electricity output for a brief period while other times remaining underutilized (Dahl, 2004). In scenarios where battery recharge could be determined based on the system load, EVs could serve as buffers, allowing for fewer and more efficient utilization of power plants (Lemoine et al., 2008). An important concern is that EVs might further exacerbate the uneven load curve if a large number of operators recharge their batteries when the system load is at its highest. This scenario would require an even more dramatic expansion in energy capacity than necessary today, which would not be used for a majority of the day, resulting in an increase of electricity prices.

Another potential impact of EVs results from the synergy between their batteries and intermittent renewable energy sources such as solar and wind power. Combining these two technologies could lead to renewables contributing a greater proportion of daily energy use because issues associated with their intermittency would be decreased. Solar cells and wind turbines could power energy storage systems such as EV batteries, which would then provide electricity as needed (Anderson, 2006). EV batteries could also complement the existing fossil-fuel based energy system by traditional power sources not having to adjust their output throughout the day, resulting in power plants being more fully utilized and lower energy prices.

1.2.4. Comparison of EVs to other alternative fuel vehicles

As identified in Section 1.1, there are several alternative fuels that have the potential to reduce GHG emissions in the transportation sector. However, pollution levels represent only one of several differences between these vehicles. Below in Table 1-16 and Figure 1-1 are technical and performance characteristics of several AFVs with an ICEV provided for base line comparison.

(20)

Table 1-17: Characteristics of midsize alternative fuel vehicles (based on US data8) Vehicle 2012 price Annual fuel costs Fuel economy9

(miles per gallon)

Fuel emissions (lbs. of CO2) Range (miles) Fueling time Fueling stations

ICEV $16,500 $1,416 26 city/36 hwy 9,605 372 4 min 121,000

EV $35,200 $600 129 city/102 hwy 7,894 73 30 min 6,806

FCV $600/mo10 $898 61 city/61 hwy 3,792 240 4 min 10

CNGV $26,305 $793 27 city/38 hwy 8,292 220 4 min 632

FFV $17,996 $1,620 20 city/28 hwy 10,464 286 4 min 2,354

HEV $18,950 $891 53 city/46 hwy 6,042 457 4 min 121,000

Because of their relatively new and complex powertrains, FCVs, HEVs, and EVs have higher purchase prices. However, due to features such as regenerative breaking, they also have the best fuel economy which leads to lower annual fuel costs. CNGVs also do well in this category because natural gas is currently cheap relative to gasoline, although its prices have shown high levels of volatility over the past 10 years (EIA, 2013). When looking at annual fuel emissions, FCVs, HEVs, and EVs (the more technologically advanced powertrains) are the best performers. In both range and fueling time, EVs stand out as performing significantly worse than the other AFVs. To further explain differences between AFVs, Figure 1-1 provides a visual representation of how they compare technologically to the core ICEV powertrain components while also noting whether the automobiles require change in fueling infrastructure.

7_{New acronyms from Table 1-1: FCV = hydrogen fuel cell vehicle, FFV = flex-fuel vehicle (uses ethanol), and}

HEV = hybrid-electric vehicle

8_{Data from the US as opposed to the EU was used for two reasons. Firstly, FCVs are not available for}

purchase/lease in the EU. Secondly, the number of fueling stations is not readily available for many EU countries.

9_{A version of this data with kilometers instead of miles is available in Appendix B.}

10_{The Honda FCX Clarity is currently only available for lease, so a purchase price comparison is not possible.} CNG

EV

FCV

None

FFV

HEV Plug-in _HEV

Changes to fueling infrastructure

Systemic Incremental Radical Changes to Core Components (ICE powertrain)

Figure 1-1: Powertrain innovations relative to the ICE and fueling infrastructure (based on figures from Henderson and Clark, 1990 and Hekkert et. al, 2005)

(21)

Some powertrains such as the FFV and CNGV still use an internal combustion engine, while others (HEV, EV, and FCV) represent a radical change to this core component through the use of batteries, electric motors, or fuel cells. In addition, AFVs also vary according to how they relate to the fueling infrastructure. For instance, the hybrid-electric powertrain runs on gasoline, so it does not require change in the current fueling system. Others AFVs including EVs and FCVs require the installation of new fueling stations. Still others such as the plug-in HEV and FFV can use existing infrastructure, but could also run on a new fuel (electricity and ethanol respectively). Regarding how Figure 1-1 relates to the development and adoption of AFVs, commercialization of an innovation becomes more difficult and expensive moving from bottom to up and left to right i.e., there are greater barriers to the introduction of FCVs than HCVs. This is a topic that will receive more attention in the subsequent section.

1.3. Theoretical factors limiting the development and adoption of EVs

Innovation literature identifies several important theoretical concepts that are particularly relevant and influential to the emergence of EVs. These include the difficulty in transitioning from a locked-in dominant design and fundamental dynamics resisting the emergence of technology such as EVs, a radical eco-innovation11 that requires a change in infrastructure and consumer behavior. These factors are detailed below.

1.3.1. Lock-in of a dominant design

Since the rise of ICEVs as the dominant automobile design almost 100 years ago, industrial dynamics have functioned to lock-in the technology as an integral part of society’s fabric, consequently erecting barriers that limit the development and adoption of competing innovations. Positive feedback through mechanisms such as learning-by-doing, economies of scale, and network externalities can serve to focus technological development along a particular path or trajectory (Dosi, 1982; van den Bergh et al., 2006). In the case of ICEVs, this has led to steady improvements in several areas including fuel efficiency, performance, safety, and comfort (Abernathy and Utterback, 1978). In addition to such incremental improvements, many dominant designs experience a buildup of supporting elements as other industries develop complementary products and services (Arthur, 1989). During the past 100 years, ICEVs have become entrenched in the fabric of everyday life through factors such as improvements in engines, expansion of fueling stations, the creation of automobile standards, and the rise of inter-industry network dependencies (Unruh, 2000). Consequently, a very strong system or what Geels (2002) refers to as a socio-technical regime has developed around the ICEV. When a technology such as the ICEV becomes dominant through technological and institutional positive feedback mechanisms, it is referred to as lock-in (Arthur, 1989). Unlocking such dominant technologies is a difficult and lengthy affair (Unruh, 2002), requiring an emerging innovation, larger macro-level changes e.g. the rise of environmentalism, and a destabilization of the existing socio-technical regime (Geels and Schot, 2007). The EV is one radical innovation that challenges the locked-in paradigm of ICEV and gasoline/diesel fuel.

1.3.2. Emergence of a radical innovation

Innovations vary in their relationship to the incumbent technology. There is a sharp distinction between those that are based on existing knowledge (incremental) and those that

11_{Following Rennings (2000), this thesis uses a broad definition of eco-innovations as the new concepts,} behavior, products, and processes, which assist in the reduction of environmental impacts or the attainment of specified ecological sustainability goals. This thesis will be mostly dealing with eco-innovations as products.

(22)

require a new source of expertise (radical) (Anderson and Tushman, 1990). In that regard, EVs represent a radical innovation because they use a high-energy battery and electric motor instead of an internal combustion engine. According to Tushman and Anderson (1986), “Major technical change opens new worlds for a product class but requires niche occupants to deal with a considerable amount of ambiguity and uncertainty as they struggle to comprehend and master both the new technology and the new competitive environment” (pg. 460). This uncertainty emerges because the extent that an innovation differs from the dominant design has an increasingly negative effect on a broad array of industrial dynamics including consumer willingness to pay, future profitability of a technology, and government involvement (Arrow, 1962; Nelson and Winter, 1977; Anderson and Tushman, 1990). And while the empirical data analyzing actions under uncertainty is “messy” (Dosi and Egidi, 1991), the theory holds that such ambiguity is a disincentive to innovation (Jaffe et al., 2005). Therefore, the radical nature of EVs increases related uncertainty and inherently acts as an obstacle to their development. Furthermore, following previous radical technologies (Adner, 2002), EVs compare poorly to ICEVs based on many traditional cost and performance metrics e.g., driving range and purchase price (see Table 1-1).

1.3.3. Lack of charging infrastructure

EV adoption faces another barrier in the lack of charging infrastructure, which is exacerbated due to the automobile’s limited driving range. Expectations regarding automobile use are based on the current paradigm where vehicles have ~375 mile (600 km) range with widely available refueling infrastructure (Egbue and Long, 2012). And while the number of charging stations has increased markedly (IEA, 2013), infrastructure shortage is still identified by consumers, auto manufacturers, and local public officials as one of the biggest challenges to wide-spread EV adoption (Egbue and Long, 2012; Zubaryeva et al., 2012). Limited charging infrastructure is often dubbed the chicken or egg problem. Consumers do not want to purchase an EV without ample available charging stations, and organizations (public and private) do not want to invest in building such infrastructure until there is a sufficiently large market (Struben and Sterman, 2008). The IEA (2013) has found national investment in charging infrastructure to be meager, especially in comparison with R&D and consumer subsidies. Financing for charging infrastructure has been identified as “perhaps the most urgent need in all EV markets” (IEA, 2013 pg. 27). As such, widespread EV adoption will require significant expansion in support infrastructure e.g., maintenance shops and charging stations in addition to appealing automobiles (Tran et al., 2013).

1.3.4. Consumer bounded rationality

Rogers (1995) noted that innovation diffusion is “an uncertainty-reduction process” (p. 232), where consumers use information about a technology when they make an adoption decision. However, this process is constrained because it is not possible for someone to have perfect information about a situation (Kahneman et al., 1986). Instead of using optimal decision making to maximize one’s utility, individuals seek only an acceptable option (Simon, 1956) because they have merely a portion of all available information (a situation referred to as bounded rationality). Consequently, the adoption of innovations is a haphazard process where the best option does not always succeed (Dosi and Nelson, 1994). Consumer bounded rationality affects EV adoption in two important ways; it often leads to misestimating lifetime ownership costs and reduces consumer willingness to pay.

In place of calculating out the total cost of ownership of a product, consumers often rely on heuristics or rules of thumb to guide their purchasing behavior (Jaffe and Stavins, 1994;

(23)

Schleich, 2009). This can lead an individual to place too much emphasis on the initial cost and not accurately value operating expenses (Levine et al., 1995). Specifically regarding EVs, consumers looking to purchase alternative fuel vehicles do not accurately incorporate fuel economy in their vehicle purchase decisions, which can lead them to buy automobiles that have a higher total life cost (Turrentine and Kurani, 2007). For these reasons, innovations that have high purchase prices and low operating expenses (such as EVs) often experience reduced rates of diffusion (Brown, 2001; Jaffe et al., 2005).

Due to limited consumer experience with EVs (by virtue of it being a new technology), information about its operation, performance, and reliability is neither well-known nor widespread (Dyerson and Pilkington, 2005; IEA, 2013). Consumer understanding of EVs is also affected by their radical differences in relation to the dominant ICEV technology. Increased uncertainty resulting from both of these factors ultimately leads to a decrease in the amount that consumers are willing to pay for EVs and consequently lower adoption rates (Arrow, 1962). One general expectation is that as consumer experience with EVs increases, then the general public’s bounded rationality regarding the innovation will go down (Mueller and Haan, 2009), increasing the likelihood that consumers will buy the automobiles. However to get to that point, it is necessary to encourage a sufficient number of early adopters to keep the market viable (Egbue and Long, 2012).

1.3.5. Eco-innovation and a pollution externality

EVs are an eco-innovation because they provide reduced environmental effects relative to gasoline or diesel fueled ICEVs, as evidenced by their lower CO2 emissions in Table 1-1. Besides helping address the environmental concerns identified in the Section 1.1, lower pollution levels also provide economic benefits such as decreased healthcare costs and fewer sick days from work as well as social benefits through improved population health and increased quality of life. However, EV adoption rates are limited because lower pollution levels are not included in the price that consumers pay. This results in pollution being an externality (a cost or benefit imposed on a third party) which can lead to market failure (the improper allocation of goods and services). As a result of this pollution externality, manufacturers are disinclined to invest in EV development because they are not compensated for all of the gains that the technology provides. In addition, environmental issues such as climate change entail such tremendous uncertainty through potential impacts and policy responses that manufacturers are disincentivized more so than normal from developing eco-innovations (Jaffe et al., 2005). According to environmental economics, public policy should be used to correct for market failure arising from pollution (Rennings, 2000).

1.4. Actors

In order to address innovation barriers such as those identified above, a broad array of actors are necessary to support both technology push (development) and demand pull (market creation) dynamics (Mowery and Rosenberg, 1979). In the case of EVs, the most important actors are auto manufactures and consumers, respectively. However, because EVs are an eco-innovation, governments also have a role to help correct for market failure arising from pollution. Furthermore, governments will also be involved because they install infrastructure such as the charging stations needed for broad EV adoption (Bakker and Trip, 2013; Egbue and Long, 2012).The roles of these three actors along with the barriers that they address are highlighted below in Table 1-2 and more specifically described in the following subsections.

(24)

Table 1-2: Role of actors relative to barriers to the development and adoption of EVs Barrier(s) addressed Actor Role Radical technology Infrastructure Bounded rationality Pollution externality

Government Address market failures X X X

Auto industry Develop EVs X X

Consumers Provide feedback on EVs X X

1.4.1. Automotive industry

In order to commercialize an EV, auto manufacturers need to acquire the necessary expertise, create a functional prototype, and then develop the production model. This is an expensive, long, and non-linear process that involves multi-directional interactions between the different innovation phases through dynamics such as learning, feedback loops, and lock-in effects (Kline and Rosenberg, 1986). The market introduction of an EV is complicated by an increase in uncertainty associated with the emergence of a radical technology (Sahal, 1981; Anderson and Tushman, 1990; Tran et al., 2013). A part of this uncertainty comes from the need for new expertise in charging, electric motors, and batteries, which are required to develop EVs. Because auto manufacturers do not have this knowledge in-house, they are frequently looking to collaborate with external organizations (Dyerson and Pilkington, 2005), which adds additional complexity to the innovation process (Powell et al., 1996).

And while EVs compare poorly to ICEVs in many cost and performance metrics, they also could have a steep improvement curve such as that seen with other radical innovations e.g., steam engines, digital storage, and personal computers (Foster, 1986; Christensen, 1997). In such a case, it is possible that battery improvements could lead to dramatic reductions in price and significant improvements in driving range such that EVs enjoy competitive advantages relative to ICEVs. Consequently, auto manufacturers may feel compelled to develop EVs because of the desire not to be left behind in the event that EVs comprise an increasing proportion of the automobile market (Dyerson and Pilkington, 2005), but related uncertainties act as a limitation on their investment in this technology.

1.4.2. Government

The primary role of public policy relative to EVs is to correct for market failure that arises from the externality pollution. As innovation policy can never be technology neutral, it always ends up favoring one particular design or another (Azar and Sandén, 2011). Thus, there is the concern that innovation policy could distort the market and ‘pick a winner’ which ends up being technologically inferior e.g., Solyndra.12 This worry is particularly pressing for alternative fuel vehicles because there are multiple competing technologies (FFVs, EVs, CNGVs, and FCVs), and support for the wrong one may lead to lock-in of an inferior technology such as that found with the QWERTY keyboard.13 In addition, the uncertainty identified above influences public policy in that it leads to policy makers having only vague notions of how an innovation’s price and performance will progress over time. Therefore, governments do not know whether support of alternative fuel vehicles, new transportation

12_{A US solar company that received more than $500 million in federal loan guarantees in 2009 before going}

bankrupt in 2011.

13_{The QWERTY keyboard is the classic innovation example where a product with superior performance (The}

(25)

modes, or efficiencies in the existing technology will be most effective in decreasing GHG emissions from the transportation sector.

Economically, taxes are the best way to correct market failures that arise from externalities such as pollution; however, they are often politically untenable (Kennedy et al., 1994; Harrington et al., 2001). So, governments have resorted to a combination of policy instruments including R&D grants to auto manufacturers, emissions standards, consumer subsidies, and charging infrastructure installation in order to support EV development and adoption (Collantes and Sperling, 2008; US DoE, 2010; ACEA, 2012a; ACEA, 2012b). A secondary role for the government is to help establish enough charging infrastructure to support wide-spread EV adoption (Tran et al., 2013). Fuel providers are hesitant to install charging stations because of the low number of EVs, and most consumers are reluctant to purchase the automobiles due to the lack of infrastructure (Struben and Sterman, 2008; Caulfield et al., 2010). Some argue that incentives or public-private partnerships are necessary to overcome this chicken or egg problem (Farrell et al., 2003). Consequently, national governments, along with some private firms and local municipalities have been investing in developing charging infrastructure to help set the stage for broader EV use (Bakker and Trip, 2013; IEA, 2013).

1.4.3. Consumers

EV price and functional capabilities have a direct influence on consumer attitudes about the innovation (Struben and Sterman, 2008; Egbue and Long, 2012). In addition to high purchase costs, consumer concerns are widespread and include a fear of being unable to find a charge station (Steinhilber et al., 2013), the long charging time (Hidrue et al., 2011; Neubauer et al., 2012), and poor performance relative to ICEVs (Lane and Potter, 2007; Hidrue et al., 2011). Furthermore, uncertainty surrounding the commercialization of EVs lowers the amount that consumers are willing to pay relative to the conventional ICEV. All of these factors results in a low number of potential EV buyers and in effect reduces their expected adoption rate (Arrow, 1962; Sovacool and Hirsh, 2009).

On the flip side, there are EV performance characteristics that make the automobiles more appealing to consumers, specifically their (potentially) low pollution emissions and the ability to achieve full torque immediately when the accelerator is depressed. As a result, EV increased acceleration capabilities have led to the development of niche markets (sports cars and eco-consumers) in which these characteristics are valued (Lane and Potter, 2007; van Bree et al., 2010). However, studies into consumer preferences show there is only a small percent of buyers that are willing to pay a premium for EVs even though they may be environmentally friendly, sporty, or innovative (Lane and Potter, 2007; Hidrue et al., 2011).

1.5. EVs 1990-2010: an abbreviated review

The technical characteristics of EVs, theoretical barriers to their market introduction, and important actors coalesced during the 1990-2010 timeframe as firms sought develop and commercialize the automobiles. Through analysis of that period, this section uses available scientific research to identify important overarching factors that have historically and continue to influence the emergence of EVs.

(26)

1.5.1. Meaningful impact

The introduction of GM’s electric concept Impact in the 1990 Los Angeles Auto Show and subsequent announcement that the automobile would be brought to production ushered in two dramatic decades for electric vehicle development and production (Bedsworth and Taylor, 2007; Dijk et al., 2013). This time saw periods of heavy commercialization activity, notably one in the 1990s and the other in late 2000s along with intermittent attention from both auto manufacturers and policy makers. However, the vast majority of consumers have estimated EVs to be unappealing mainly due to their high costs and limited performance capabilities, resulting in a failed commercial attempt in the 1990s, and a significant barrier to their introduction in the late 2000s (Dijk and Yarime, 2010; IEA, 2013).

Governments have introduced a wide range of policies to encourage EVs diffusion. Notably, the California Air Resource Board’s low emissions vehicle program in 1990 mandated the sale of zero emissions vehicles later in that decade (Collantes and Sperling, 2008). However, that US state was not alone; countries around the world have implemented supportive policies. European nations encouraged the introduction of EVs largely through R&D programs and pilot projects. For example, almost 400 EVs were employed through a demonstration effort in the Swiss town Mendriso, and 2,000 of the automobiles were the target of an extensive field test in the French city La Rochelle (Hoogma, 2002). The Japanese Ministry of International Trade and Industry issued an aggressive market expansion policy in 1991, followed by a series of pilot projects throughout the 1990s (JEVA, 2000). These programs had the goal of putting hundreds of thousands of EVs on the road by the 2000s (MITI, 1990; Bedsworth and Taylor, 2007; Hoogma, 2002).

During this period, incumbent auto manufacturers invested tremendous resources in developing EVs, and produced several prototype models including Ford’s Ecostar, Honda’s EVX, BMW’s e1, and the Nissan FEV-I (Mom, 1997). Only a few of these prototypes were ever introduced to the market as production models, notably GM’s EV1 and Toyota’s RAV4EV, and by the early 2000s, manufacture of EVs had practically stopped (Dijk et al., 2013). These automobiles suffered from the same barriers to adoption as the current wave of EVs (high purchase price, low driving range, and little charging infrastructure). Although their situation was exacerbated because they were using lower energy density battery technology (lead-acid or nickel based), resulting in even lower driving ranges. Primarily due to production costs, EVs were determined not to represent a viable business model; pilot projects ended, supportive policies were severely watered down, and auto makers gradually retreated from the EV market in the early 2000s (Patchell, 1999; Funk and Rabl, 1999; Dijk and Yarime, 2010).

1.5.2. Retrenchment and re-emergence

After interest in EVs died down, focus shifted to different low-emissions vehicle powertrains including HEVs and FCVs (Dijk and Yarime, 2010; Bakker et al., 2012b). Governments specifically supported these AFVs through policies such as FreedomCAR in the US and the Clean Energy Vehicles Introduction Program in Japan. Auto manufacturers also devoted resources toward developing expertise in those powertrains (Oltra and St. Jean, 2009). And while the HEV can be seen as a commercial success (Dijk and Yarime, 2010), FCVs failed to live up to expectations, following EVs into disappointment along Gartner’s technology hype cycle (Bakker, 2010).

Convergence of a series of factors including more stringent fuel emissions legislation, supportive R&D policies, improvement in battery technology, and higher fuel prices

(27)

contributed to the re-emergence of EVs in the late 2000s. The wide-scale commercialization of the Nissan LEAF and Mitsubishi iMiEV in 2009 along with the appearance of startup manufacturers such as Coda and Tesla indicated a level of momentum behind the most recent introduction of EVs. And while there is disagreement about its future success prospects (Dijk et al., 2013; Wells and Nieuwenhuis, 2012), EVs have reached a level of commercialization much greater than that found in the earlier 1990’s attempt. However, a general conclusion about this development is that any sort of broad EV diffusion in the future will still require supportive governmental policy, industrial buy-in, and changes in consumer behavior (van Bree et al., 2010; Tran et al., 2013; Dijk et al., 2013). In one positive sign, recent market introductions indicate that large auto makers now view the EV market as a commercial opportunity instead of a regulatory requirement (Magnusson and Berggren, 2011), specifically in niche markets such as sports cars and low emissions vehicles (van Bree et al., 2010). And while auto manufacturers have a diverse patent portfolio of automotive technologies e.g., FCVs, and HEVs (Oltra and St. Jean, 2009), the firms are beginning to show more of a preferential attitude toward EVs (Schwedes et al., 2012).

Increasingly stringent environmental policies, notably the 2009 US fuel economy standards and 2009 EU vehicle emission regulations, have affected EV commercialization in two important ways. Firstly, they encourage auto makers to sell EVs since it helps them meet regulatory requirements. Conversely, they reduce EV operational advantages because they result in lower ICEV fuel costs. However, improved fuel economy and lower emissions could come through increased manufacturing costs and subsequently a higher purchase price, causing an improvement in the EV/ICEV value proposition. Thus, EVs will have to contend with ICEVs that are steadily improving in their operational costs. This dynamic is often present when a radical technology offers a new price/performance frontier and functions to slow adoption rates (Geels, 2002).

1.5.3. Summary of this period

Based on studies during the 1990-2010 timeframe, EV adoption is seen as being very limited without stimulation from external factors such as stringent emissions regulations, rising fuel prices, or financial incentives (Eppstein et al., 2011; Shafei et al., 2012; Tran et al., 2013). Of those factors, consumer subsidies in particular are expected to be necessary for EVs to reach a mass market (Hidrue et al., 2011; Eppstein et al., 2011). Subsequently, governments around the world have implemented this demand-pull instrument through different types of financial incentives (IEA, 2011; IEA, 2013). Countervailing forces include consumer uncertainty regarding new technological components and operation as well as the gradual improvement in ICEV fuel efficiency. If consumer confidence in a technology is lacking, then financial incentives will not be very effective in stimulating EV diffusion (Egbue and Long, 2012). A common conclusion that policy makers, researchers, and auto manufacturers draw about the current prospects for EV commercialization is that they are uncertain what the future of the innovation will be (Egbue and Long, 2012; Tran, 2013; IEA, 2013).

1.6. Research gap and question

The above section gives a literature overview which explains the current understanding of factors which influence the development and commercialization of EVs. Building on that foundation, this section of the thesis identifies a hole within the literature and an associated research question used to bridge this gap, providing a better grasp of important dynamics that impact the emergence of EVs.

(28)

There is the concern that the existing EV literature may not accurately reflect the current industrial environment because most studies analyzed stated as opposed to observed consumer behavior, or were conducted before the most recent commercialization of the automobiles. Due to the value-action gap, stated preference surveys may not correctly identify consumer behavior regarding EVs. As a result, there are reasons to doubt whether studies using such surveys correctly reflect consumer attitudes toward EVs (Homer and Kahle, 1998; Lane and Potter, 2007). And while the influence of government policies on consumer adoption of EVs has been studied with agent based modeling (Epstein et al., 2011; Shafei et al., 2012), that does not provide the sort of insight or certainty that comes from empirical analysis, which is now possible that the automobiles have been available for purchase for several years. In addition, because industrial dynamics change so quickly during the emergence of a radical innovation (Tushman and Anderson, 1986; Klepper, 1996), studies focusing on manufacturer activities either need to be updated (Oltra and St. Jean, 2009), or expanded to look at a broader set of firms (Magnusson and Berggren, 2011).

The primary research gap which this thesis seeks to fill is the lack of knowledge regarding recent efforts of the automotive industry to develop and commercialize EVs. While earlier studies focused on the R&D stage of EV development, the innovation has moved on to the commercialization phase, identifying the need for an updated understanding of the industrial dynamics at work. And because EVs have been broadly available for purchase for a number of years, it is now possible to empirically analyze important factors such as prototype and production model development, alliance formation, and vehicle sales. As such, the central research question of this thesis is:

How has the automotive industry approached the development and commercialization of electric vehicles?

To answer that question, this thesis uses a series of sub-queries which each occupy a single chapter of this thesis. Figure 1-2 below shows how these chapters are positioned relative to one another and the broader EV industry, while the following subsections describes the analysis that was conducted for each chapter.

Figure 1-2: Thesis chapter positioning within the broader EV industry Alternative fuel vehicle development ③

Electric vehicle industry ④

Automobile manufacturers Incumbent s Startups Related industries e.g., batteries and electric motors Government Consumers ⑥ ⑤ ②

(29)

1.6.1. Knowledge creation

As one of the characteristics of radical technical change is that new innovations require expertise outside that necessary for the conventional technology (Anderson and Tushman, 1990), auto firms have been rapidly trying to accumulate knowledge in fields such as batteries and electric motors (Dyerson and Pilkington, 2005; Magnusson and Berggren, 2011). Based on patent research, auto manufacturers have been actively developing their own knowledge regarding those key electric vehicle technologies (Oltra and St. Jean, 2009; Wesseling et al., 2013). However, the increased complexity of technologies means that firms are often no longer able to develop radical innovations on their own. Powell et al. (1996) have determined that the locus of innovation has shifted away from individual firms and toward networks of organizations. Therefore, to understand how auto manufacturers are acquiring the expertise necessary to develop EVs, it is important to analyze the collaborations they are making in key knowledge areas. The primary research question for Chapter 2 is how have auto manufacturers approached the acquisition of knowledge from disparate industries in order to produce a commercial electric vehicle?

1.6.2. Alternative fuel vehicle introduction

Within the broader auto industry, manufacturers have commercialized (or are attempting to commercialize) many alternative fuel powertrains including hydrogen fuel cells, hybrid-electric, purely hybrid-electric, and engines that can run on CNG and biofuel (Oltra and St. Jean, 2009; Dijk and Yarime, 2010; Yu et al., 2010; Bakker et al., 2012). These alternative fuel vehicles provide different technological approaches for lowering emissions in the transportation sector. To understand the early adoption environment of EVs, it is necessary to see how that particular powertrain fits into a broader market for alternative fuel vehicles. The primary research question for the third chapter is, how have incumbent auto firms approached the development of electric automobiles relative to other alternative fuel vehicles?

1.6.3. The emerging electric vehicle market

There are several different dynamics which indicate that a radical technological shift could be underway in an industry including an increase in technological variety, more startups, and heightened uncertainty (van Dijk, 2000; Klepper, 1996). While Chapter 3 gives a broad overview of the development of alternative fuel technologies, it does not provide much detail for what is happening specifically in the EV market. A more in-depth analysis into that area is supplied in Chapter 4 which looks at the important industrial dynamics of technological variety and startup vs. incumbent firm behavior. The primary research question for this section is, to what extent did incumbent and startup firms develop a variety of different electric vehicle types based on performance criteria?

1.6.4. Consumer financial incentives and EV adoption

In general, governments are using broad emissions regulation to gradually improve the environmental impact of vehicles, while being more selective over their use of technology-specific policies. Uncertainty about the EV industry has made it difficult for policy makers to determine if and how to support the technology (Struben and Sterman, 2008; van Bree et al., 2010; Tran et al., 2013). There are several policy measures available including consumer financial incentives, infrastructure development, producer subsidies/loans, and emissions regulation. Historically, these measures have had a mixed success rate e.g., HEV adoption, loan recipients Tesla and Solyndra, and ZEV/CAFE regulation (Bedsworth and Taylor, 2007; Diamond, 2009; Gallagher and Muehlegger, 2011). Because EVs have only been widely

(30)

available since approximately 2010, there is little data to know how the policies that are in place have fared. Consequently, much of the research looking into the effectiveness of EV policies has relied on surveys from the general public (not from adopters) (Eppstein et al., 2011; Egbue and Long, 2012; Hidrue et al., 2011). However, because of a phenomenon known as the ‘value-action gap’ there is the concern that information from consumer surveys may have little relation to the purchase of cleaner vehicles (Lane and Potter, 2007). As such, governments are implementing policies without a clear understanding of their effectiveness. The fifth chapter of this thesis focuses on the relationship between financial incentives and EV adoption to help note how governments could help stimulate diffusion of the innovation. Here the research question is to what extent do consumer financial incentives and other socio-economic factors explain national EV adoption rates?

1.6.5. Fleet manager adoption of EVs

Consumers often reject new technologies and instead rely on a notion of tradition or familiarity when considering products, especially for hardware (such as an automobile) that has high capital costs (Rogers, 1995; Kirsch, 2000). Because the current commercialized EVs have only been on the market for a few short years, there is little available data on their reliability and safety. As such, the vehicles have not been on the road long enough to be considered ‘tested’ (BERR, 2008). Because of these issues, the public is unfamiliar with EVs which discourages consumer adoption (Sovacool and Hirsh, 2009). Therefore the reasons why some consumers have adopted EVs needs to be identified to better understand the demand side of the market.

One of the difficulties limiting the early adoption of a radical innovation such as electric vehicles (EVs) is the capture of a receptive consumer market (Christiansen, 1997). The literature has identified several reasons why fleet managers are good candidates to be EV early adopters such as their intense usage and high automobile purchase rates. This expectation is supported by a recent report from Frost and Sullivan (2013) which found that to 2013, governments and firms have been responsible for a majority of EV purchases. The research question of this sixth chapter is, what were the important factors that influenced fleet managers’ initial adoption of EVs?

1.7. Scope, data, and methods

This section identifies the research scope, data, and methods that were used for analysis. The complexity and breadth of the EV innovation process results in it being much too large to address in a single study. As such, the scope of this thesis was limited primarily to the role of auto manufacturers during EV market introduction with attention also devoted to consumer financial incentives and early adopters. While some chapters (3 and 4) deal with the broader timeframe of 1991-2011, a majority of the analyses focus on the recent buildup of EV expertise and the introduction of production models since approximately 2007. This approach allows the thesis to concentrate on recent actions and dynamics within the EV industry. Because innovation is not confined within a country’s borders and auto manufacturers are multi-national corporations, these studies generally take a global perspective, although Chapter 6 examines early adopters from the Netherlands and US.

Since a goal of this thesis is to study the emerging state of the EV industry, the constantly changing market environment creates data issues because information needs to be both current and reliable. Accordingly, analyses herein depend on up-to-date data e.g., vehicle prototypes, public charging station maps, and inter-firm alliances. In this regard, individual

(31)

studies make use of proven collection and analysis methods when dealing with publicly available information.

This thesis employed both inferential and descriptive analytical methods, including content analysis (both qualitative and quantitative), linear regression using ordinary least squares, t-tests, and frequency distributions. Individual chapters 2-6 provide more specific detail about the methods used for each analysis.

(32)

References

Abernathy W. and J. Utterback (1978) Patterns of Industrial Innovation, in: Technology Review, 80, 41-47.

ACEA (2012a) Overview of CO2 based motor vehicle taxes in the EU. Brussels, Belgium. ACEA (2012b) Overview of tax incentives for electric vehicles in the EU. Brussels, Belgium. Adner R. (2002) When are Technologies Disruptive? A Demand-Based View of the Emergence of Competition, in: Strategic Management Journal, 23, 667-688.

Anderson D. (2006) Costs and finance of abating carbon emissions in the energy sector. Cambridge, UK, A report prepared for the HM Treasury Stern Review on the economics of climate change.

Arrow K. (1962) The Economic Implications of Learning by Doing, in: Review of Economic Studies, 29(3), 155-73.

Arthur B. (1989) Competing Technologies, Increasing Returns, and Lock-In by Historical Events, in: The Economic Journal, 394(99), 116-131.

Autotrader (2013) 2013 Nissan Leaf vs. 2013 Ford Focus Electric. http://www.autotrader.com/research/article/car-reviews/212049/2013-nissan-leaf-vs-2013-ford-focus-electric.jsp [Accessed 9.27.2013].

Azar C. and B. Sandén (2011) The elusive quest for technology-neutral policies, in: Environmental Innovation and Societal Transitions, 1, 135-139.

Bakker S. (2010) The care industry and the blow-out of the hydrogen hype, in: Energy Policy, 38(11), 6540-6544.

Bakker S., H. van Lente, and R. Engels (2012a). Competition in a technological niche: the cars of the future, in: Technology Analysis & Strategic Management, 24(5), 421-434.

Bakker S., H. van Lente and M. Meeus (2012b). Dominance in the Prototyping Phase-the Case of Hydrogen Passenger Cars, in: Research Policy, 41(5), 871-883.

Bakker S. and J. Trip (2013) Policy options to support the adoption of electric vehicles in the urban environment, in: Transportation Research Part D, 25, 18-23.

Bedsworth L. and M. Taylor (2007) Pushing technology when it pushes back: learning from California’s zero-emission vehicle program. California Economic Policy Report, Public Policy Institute of California.

Boulanger A., A. Chu, S. Maxx, and D. Waltz (2011) Vehicle Electrification and Issues, in: Proceedings of the IEEE, 99(6), 1116-1138.

Brown M. (2001) Market failures and barriers as a basis for clean energy policies, in: Energy Policy, 29(14), 1197-1207.

(33)

Caulfield B., S. Farrell and B. McMahon (2010) Examining individuals preferences for hybrid electric and alternatively fuelled vehicles, in: Transport Policy, 17, 381-387.

Christensen C. (1997) The Innovator’s Dilemma, Harvard Business School Press, Boston, MA.

Collantes G. and D. Sperling (2008) The origin of California's zero emission vehicle mandate, in: Transportation Research Part A, 42, 1302-1313.

Diamond D. (2009) The impact of government incentives for hybrid-electric vehicles: Evidence from US states, in: Energy Policy, 37, 972-983.

Dijk M. and M. Yarime (2010) The emergence of hybrid-electric cars: innovation path creation through co-evolution of supply and demand, in: Technological Forecasting and Social Change, 77, 1371-1390.

Dijk M., R. Orsato and R. Kemp (2013) The emergence of an electric mobility trajectory, in: Energy Policy, 52, 135-145.

Dosi G. (1982) Technological paradigms and technological trajectories, in: Research Policy, 11, 147-162.

Dosi G. and M. Egidi (1991) Substantive and procedural uncertainty, in: Journal of Evolutionary Economics, 1, 145-168.

Dosi, G. and R. Nelson (1994) An introduction to evolutionary theories in economics, in: Journal of Evolutionary Economics, 4, 153-172.

Dyerson R. and A. Pilkington (2005) Gales of creative destruction and the opportunistic incumbent: The case of electric vehicles in California, in: Technology Analysis & Strategic Management, 17(4), 391-408.

Egbue O. and S. Long (2012) Barriers to widespread adoption of electric vehicles: an analysis of consumer attitudes and perceptions, in: Energy Policy 48, 717-729.

EIA (2013) Henry Hub Gulf Coast Natural Gas Spot Price. http://www.eia.gov/dnav/ng/hist/rngwhhdm.htm [Accessed 12.4.2013].

Eppstein M., D. Grover, J. Marshall, and D. Rizzo (2011) An agent-based model to study market penetration of plug-in hybrid electric vehicles, in: Energy Policy, 39, 3789-3802. Farrell A., D. Keith and J. Corbett (2003) A strategy for introducing hydrogen into transportation, in: Energy Policy, 31, 1357-1367.

Foster R. (1986) Innovation: The Attacker’s Advantage, Macmillan, London. Frost and Sullivan (2013) Strategic Outlook of the Global Electric Vehicle Market.

Funk K. and A. Rabl (1999) Electric versus conventional vehicles: social costs and benefits in France, in: Transportation Research Part D, 4, 397–411.

Gallagher K. and E. Muehlegger (2011) Giving green to get green? Incentives and consumer adoption of hybrid vehicle technology, in: Journal of Environmental Economics and Management, 61(1), 1-15.

Geels F. (2002) Technological transitions as evolutionary reconfiguration processes: a multi-level perspective and a case-study. Research Policy, 31(8–9), 1257–1274.

Harrington W., A. Krupnick and A. Alberini (2001) Overcoming public aversion to congestion pricing, in: Transportation Research Part A, 35(2), 87-105.

Development and Early Adoption of Electric Vehicles: Understanding the tempest

Preface

Table of Contents

L

ist

of Figures

List of Tables

1.

Introduction

1.1.

Uncertainty in the automotive sector

1.2.

Technical overview of electric vehicles

1.3.

Theoretical factors limiting the development and adoption of EVs

1.4.

Actors

1.5.

EVs 1990-2010: an abbreviated review

1.6.

Research gap and question

1.7.

Scope, data, and methods

References